Parallelization and cross-correlation of change detection using multi-temporal, multi-source satellite imagery

点击翻译按钮获取中文摘要

Biosphere Management Systems Inc. (BMSI), in association with Urthecast Inc. (UrtheCast), is a Vancouver-based company specializing in commercial earth observation systems. Currently, BMSI is seeking to design and develop an interactive software system that (i) provides users with the ability to monitor certain geographical locations in terms of detecting change using multi-temporal satellite imagery, and (ii) leverages modern computing paradigms such as cloud computing through parallelization of tasks among a vast number of computational units to deliver a fast and efficient user experience. BMSI has approached the University of Victoria to design and develop a working prototype solution that meets these objectives. In this context, this project involves selecting a set of change detection techniques that are suitable for BMSI's problem domain, designing and developing a software that provides the functionality of the selected techniques in a way that supports task parallelization, and investigating how correlations among the results of different change detection techniques can be used to represent the level of confidence in the detected change. The prototype solution will consist of a web-based front-end where the users can interact with the system in order to view the status (change/no change) of a geographical location at different instances, and a back-end system that supports parallelization of change detection techniques and cloud deployment.